Informed Trading When Information Becomes Stale


  • Dan Bernhardt,

  • Jianjun Miao

    Search for more papers by this author
    • Bernhardt is with the Department of Economics, University of Illinois at Urbana-Champaign. Miao is with the Department of Economics, Boston University. We thank Shmuel Baruch, Oleg Bondarenko, Larry Epstein, Terrence Henderschott, Eric Hughson, Leslie Marx, and a particularly conscientious referee for helpful comments. We are especially grateful to Motohiro Sato. Our analysis was tremendously facilitated by Motohiro Sato's extended solution to a homework problem in a market microstructure finance course at Queen's University in 1997.


This paper characterizes informed trade when speculators can acquire distinct signals of varying quality about an asset's value at different dates. The most reasonable characterization of private information about stocks is that while information is long-lived, new information will arrive over time, information that may be acquired by others. Hence, while a speculator may know more than others at a moment, in the future, his information will become stale, but not valueless. In an environment that allows for arbitrary correlations among signals, we characterize equilibrium outcomes including trading, prices, and profits. We provide explicit numerical characterizations for different informational environments.